arXiv：cs.LG· Feiqing Huang, Zongqi Xia, Rong Ma, Tianxi Cai·· 4 小时前AI 评分33
通过稳健灵活的知识迁移增强电子健康记录中的谱嵌入
Enhancing Spectral Embedding through Robust and Flexible Knowledge Transfer in Electronic Health Records
AI 导读
研究提出一种基于谱的无监督表示学习框架，从电子健康记录中为罕见病队列的临床概念和患者生成低维嵌入，并引入来自更广泛人群的知识矩阵。该方法放宽了潜在数据矩阵与知识矩阵之间严格的一对一信号对齐假设，采用两步谱嵌入：先剔除知识矩阵中的无关成分，再用基于投影的方法分别恢复共享与异质成分。仿真与真实多发性硬化症队列分析显示，其在共享信号微弱且仅部分对齐的场景下优于对比方法。
正文
Abstract:We propose a spectral-based, unsupervised representation learning framework to derive low-dimensional embeddings for clinical concepts and patients in rare disease cohorts from electronic health records, where data are high-dimensional but sample sizes are limited. To overcome this challenge, we incorporate a knowledge matrix extracted from a broader population that shares a partially overlapping subspace with the rare-disease cohort. Our method departs from existing approaches by relaxing restrictive one-to-one signal-alignment assumptions between the latent data matrix and knowledge matrix, allowing more flexible and realistic forms of structured sharing. We introduce a novel two-step spectral embedding procedure: first, we identify and remove irrelevant components from the knowledge matrix; then, we apply a projection-based method to separately recover shared and heterogeneous components. Simulations and an analysis of a real-world multiple sclerosis cohort show that the proposed method outperforms competing approaches, particularly in challenging scenarios where shared signals are weak and only partially aligned, as is common in rare-disease data.
| Subjects: | Machine Learning (stat.ML); Machine Learning (cs.LG); Methodology (stat.ME) |
| Cite as: | arXiv:2606.11570 [stat.ML] |
| (or arXiv:2606.11570v2 [stat.ML] for this version) | |
| https://doi.org/10.48550/arXiv.2606.11570 arXiv-issued DOI via DataCite |
Submission history
From: Feiqing Huang [view email]
[v1]
Wed, 10 Jun 2026 01:51:51 UTC (7,042 KB)
[v2]
Mon, 5 Oct 2026 18:40:48 UTC (7,033 KB)
来源：arXiv：cs.LG · arxiv.org